Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Sat, 10 Sep 2022 15:20:35 +0100
Message-ID: <871qsjqpjg.wl-maz@kernel.org>
From:   Marc Zyngier <maz@kernel.org>
To:     Will Deacon <will@kernel.org>
Cc:     Elliot Berman <quic_eberman@quicinc.com>,
        Catalin Marinas <catalin.marinas@arm.com>,
        James Morse <james.morse@arm.com>,
        Alexandru Elisei <alexandru.elisei@arm.com>,
        Suzuki K Poulose <suzuki.poulose@arm.com>,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
        kvmarm@lists.cs.columbia.edu
Subject: Re: [PATCH v2] KVM: arm64: Only set KVM_MODE_PROTECTED if is_hyp_mode_available()
In-Reply-To: <20220910134342.GA959@willie-the-truck>
References: <20220909144552.3000716-1-quic_eberman@quicinc.com>
        <Yxt3wmXYYbWraXrd@arm.com>
        <05057e2a-1a85-69ba-ffcd-584d4090467a@quicinc.com>
        <878rmrr3xw.wl-maz@kernel.org>
        <20220910134342.GA959@willie-the-truck>
User-Agent: Wanderlust/2.15.9 (Almost Unreal) SEMI-EPG/1.14.7 (Harue)
 FLIM-LB/1.14.9 (=?UTF-8?B?R29qxY0=?=) APEL-LB/10.8 EasyPG/1.0.0 Emacs/27.1
 (x86_64-pc-linux-gnu) MULE/6.0 (HANACHIRUSATO)
MIME-Version: 1.0 (generated by SEMI-EPG 1.14.7 - "Harue")
Content-Type: text/plain; charset=US-ASCII
Precedence: bulk

On Sat, 10 Sep 2022 14:43:44 +0100,
Will Deacon <will@kernel.org> wrote:
> 
> On Sat, Sep 10, 2022 at 10:09:31AM +0100, Marc Zyngier wrote:
> > On Fri, 09 Sep 2022 18:55:18 +0100,
> > Elliot Berman <quic_eberman@quicinc.com> wrote:
> > > 
> > > 
> > > 
> > > On 9/9/2022 10:28 AM, Catalin Marinas wrote:
> > > > On Fri, Sep 09, 2022 at 07:45:52AM -0700, Elliot Berman wrote:
> > > >> Do not switch kvm_mode to KVM_MODE_PROTECTED if hypervisor mode is not
> > > >> available. This prevents "Protected KVM" cpu capability being reported
> > > >> when Linux is booting in EL1 and would not have KVM enabled.
> > > >> 
> > > >> Signed-off-by: Elliot Berman <quic_eberman@quicinc.com>
> > > >> ---
> > > >>   arch/arm64/kvm/arm.c | 4 +++-
> > > >>   1 file changed, 3 insertions(+), 1 deletion(-)
> > > >> 
> > > >> diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
> > > >> index 8fe73ee5fa84..861f4b388879 100644
> > > >> --- a/arch/arm64/kvm/arm.c
> > > >> +++ b/arch/arm64/kvm/arm.c
> > > >> @@ -2272,7 +2272,9 @@ static int __init early_kvm_mode_cfg(char *arg)
> > > >>   		return -EINVAL;
> > > >>     	if (strcmp(arg, "protected") == 0) {
> > > >> -		if (!is_kernel_in_hyp_mode())
> > > >> +		if (!is_hyp_mode_available())
> > > >> +			kvm_mode = KVM_MODE_DEFAULT;
> > > > 
> > > > I think kvm_mode is already KVM_MODE_DEFAULT at this point. You may want
> > > > to print a warning instead.
> > > > 
> > > 
> > > Does it make sense to print warning for kvm-arm.mode=nvhe as well?
> > 
> > In general, specifying a kvm-arm.mode when no hypervisor mode is
> > available should be reported as a warning.
> 
> As long as this is pr_warn() rather than WARN() then I agree. Otherwise,
> kernels with a kvm-arm.mode hardcoded in CONFIG_CMDLINE (e.g. Android's
> GKI) will make for noisy guests.

Indeed, pr_warn() is what I had in mind. A WARN() would be pretty
overkill, as there is nothing majorly wrong with booting at EL1, just
an impossibility to honour the request from the command line.

	M.

-- 
Without deviation from the norm, progress is not possible.